COVID, Clots and Platelets

COVID, Clots and Platelets

The very rare occurrence of death due to "blood clots following vaccination by AstraZenica's anti SARS-CoV-2 vaccine” (hereafter BCfAZV) is rightly causing concern. It is quite hard to get these rare events into perspective. Not from want of information; the authorities are being exemplary in the amount and clarity of the information they publish. No; it is more the amount of information, its complexity, and the repetitive nature of media coverage laced with strange medical terms that wears you down. And the very “rarity” of rare events is hard to grasp.

Platelets

We are told that these vaccine-induced clots are of a rare type — showing low or very low platelet count. Platelets (also called ’thrombocytes’) are blood cells charged with the job of clotting, and blocking the leakage of blood from blood vessels [1]. Too few platelets (a condition called thrombocytopenia) might indicate potential failure to clot; as when taking too much Warfarin. But in the thrombocytopenia following viral infection it seems that ‘something’ has triggered platelet-activation; the platelets cluster, stick, and die. Clusters of dead/dying platelets can break free and flow in the blood till they reach a narrowing, and block the flow. That is the danger. The remaining blood is depleted of thrombocytes/platelets.

Virus-induced Thrombocytopenia 

A somewhat lowered platelet count (< 100 × 10^6/mL blood) has been observed following a number of viral infections, including hepatitis B (and C) viruses, cytomegalovirus, Varicella zoster virus, HIV, and the arboviruses zika and dengue. In this latest COVID-19 pandemic it is found in up to one-third of COVID  patients, so we can add wild-type SARS-CoV-2 to that list of viruses that induce thrombocytopenia. 

One suggested explanation of the vary rare occurrence of BCfAZV is that after vaccination one or two patients go on to contract COVID, though (to date 2021-04-17) the diagnostic viral RNA has not been detected by PCR [2 (page 37)].

Abnormal clotting can be a factor in the pathology of pandemic and seasonal ‘flu (due to influenza A(H1N1) and other viruses)[3]. In these cases the mechanism may be different, may involve proteases, and pre-disposing genetic polymorphisms in complement proteins. But I flag it because there are a few cases, for that virus disease also, where it seems that it was vaccination that triggered the activation of the platelets.

Rare Events

An early response to the suggestion that the AstraZeneca COVID-19 vaccine causes blood clots was to point out that clots do occur at a similar frequency without vaccination [4,Thromboembolism and the Oxford–AstraZeneca COVID-19 vaccine: side-effect or coincidence?]. The annual incidence of cerebral venous sinus thrombosis is said to be between 2 and 5 per million people [5]. Recent data (as of 4th April 2021) from the 34 million people who have just received the AstraZeneca COVID-19 vaccine are: 169 cases of thrombus in the cerebral venous sinus, and 53 with thrombi in the splanchnic vein. If we restrict ourselves to the cerebral venous sinus, that is 4.97 per million. An observed/expected ratio of 4.97/ 5 is obviously not significant; until you realise that these blood clots occurred within 7 - 30 days of vaccination. So the observed rate could be 12 - 40 times the expected rate.

          Some commentators have tried to picture the “5-in-a-million” risk by talking of the risk of a fatal traffic accident on our roads, though that is yet another complicated question. In 2019 there were 1752 fatalities in the UK (population=67 million), which corresponds to 26.5 deaths per million citizens per year [11]; though road-death risk is clearly lower for some and therefore higher for others. 

To what new antigens are vaccinees exposed?

The RNA vaccines expose the recipient to the expressed SARS-CoV-2 spike protein and to phospholipid; that is to say, they are rather “pure”.

The AstraZenica vaccine carries DNA for the viral spike protein, together with coding for a 36 aa portion of tissue plasminogen activator leader sequence, plus DNA for all the components of the Chimpanzee Adenovirus vehicle (ChAdOx-1), plus the Adenovirus proteins themselves. (The choice of a replication-incapable Chimpanzee virus instead of a human strain as the vehicle was presumably so that it would be very unlikely that the recipient would already have circulating antibodies against the inoculum.) 

The Johnson & Johnson (= Janssen) vaccine uses human Adenovirus HAdV-D26 as vehicle. (There are over 80 different strains of human Adenovirus known, grouped into species (A, B, C, D, etc.)) Fewer doses of this Janssen vaccine than of the AstraZenica one have been administered to date, but a case of abnormal clotting following injection has now been reported [6].  

Human Adenovirus vehicles have been used since 2000 in exploratory experiments on gene therapy, and quite a lot of the resulting interactions between virus and host are well known. Thus, it is known that Adenovirus injected into a blood vessel rapidly binds to platelets [7], and coagulation factors in the blood [8][9]. 

Why, then, is fatal clotting so rare?

If Adenovirus vehicles normally interact with platelets and coagulation factors, why is the problem of thrombocytopenia and clots-in-veins so rare? One possibility is that there may be a rare genetic polymorphism in the human population that predisposes carriers of that rare genotype to full-blown platelet activation, and clots [10]. A second suggestion is that, when a nurse is giving 300 intramuscular injections in a morning, an occasional needle might pierce a blood vessel, and administer an intravenous injection by mistake [10].  

        The AstraZenica vaccination is still safer than road travel.

References

[1] https://www.verywellhealth.com/thrombocyte-what-is-a-thrombocyte-797228

[2] https://www.ema.europa.eu/en/documents/prac-recommendation/signal-assessment-report-embolic-thrombotic-events-smq-covid-19-vaccine-chadox1-s-recombinant-covid_en.pdf

[3] Pediatr Nephrol. 2018; 33(11): 2009–2025.

[4] The Lancet, Vol. 397, Issue 10283, pp.1441-1443, April 17, 2021.

[5]  https://www.ema.europa.eu/en/documents/prac-recommendation/signal-assessment-report-embolic-thrombotic-events-smq-covid-19-vaccine-chadox1-s-recombinant-covid_en.pdf  

[6] DOI: 10.1056/NEJMc2105869

[7] https://jvi.asm.org/content/81/9/4866

[8] https://febs.onlinelibrary.wiley.com/doi/10.1002/1873-3468.13649

[9] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4009923/

[10] https://www.ema.europa.eu/en/documents/prac-recommendation/signal-assessment-report-embolic-thrombotic-events-smq-covid-19-vaccine-chadox1-s-recombinant-covid_en.pdf 

[11] https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/922717/reported-road-casualties-annual-report-2019.pdf

(Please comment directly to cawstein@gmail.com) 

 SARS-CoV-2: Variants of Concern

            The definitive reference sequence of the SARS-CoV-2 RNA is that of the Wuhan strain, isolated from a patient admitted to Hospital in Wuhan on 26th December 2019, published in Nature [5]. It is no more important than any of the other closely related sequences. Currently it is not established that this sequence has a seminal status, from which other human strains were in fact derived, but because it is the arbitrarily chosen reference strain, it will seem to be seminal. 

            Single-stranded RNA is notoriously susceptible to mutation (which may be why 'higher' forms of life settled on double-stranded DNA to carry genetic information).

            As early as January 2020 a variant was sequenced in Germany in which the amino acid at position 614 of the Spike protein was not Aspartate (D) as in the Wuhan strain but was Glycine (G); thus described as a D614G mutation [4]. This form spread rapidly among humans and by the summer was found in 97.5% of isolates world-wide. In December this mutation was shown to cause tighter binding to the ACE-2 receptor and higher infectivity [3] than the Wuhan reference strain. However, it is now essentially ubiquitous in humans, to whom it seem advantageously adapted (in the Darwinian sense). Further adaptations to humans are to be expected in the human population. No doubt 'cat' variants are arising and spreading in cat populations, but cat travel is limited in comparison with human.

            The GISAID repository, set up in 2008 as a place where flu virus sequences could be housed and shared, now (2020-12-20) contains some 270,000 SARS-CoV-2 sequences, of which 120,000 were contributed by the UK consortium "Covid-19 Genomics UK (COG-UK)".  There are currently some 4,000 mutations in the spike gene alone [7]. Many are 'silent'; not every mutation in the RNA shows in the protein as most amino acids have several alternative codons, so giving 'synonymous' mutations. Many are trivial, in that an amino acid may be replaced by a similar one; so 'conservative' mutations. 

            Two publications from the European Centre for Disease Control (ECDC) are useful [1,6].  There are now two "variants of concern" (VOC): the rapid-spreader that saw Britain cut off from Europe on 21 December, named VOC 202012/01; and a rapid-spreading variant in South African called 501.V2.

            The British rapid-spreading variant was first sequenced in October [1,7], but may have occurred earlier. As of 2020-12-26 there have been 3000 cases of VOC 202012/01 identified in UK by sequencing. Throughout December the UK has been sequencing some  3000 – 7000 COVID genomes a week. From being a rarity (<0.1%) the variant has grown exponentially  to become 11% of the genomes sequenced by the end of November, roughly doubling its market share each week [1]. In Norfolk it accounts for 20% of COVID cases [7]

            VOC 202012/01 is defined by nine spike protein mutations (deletion 69-70, deletion 144, N501Y, A570D, D614G, P681H, T716I, S982A, D1118H), and 8 mutations in other genomic regions [1,7]. The D614G  substitution (see above) is present. The P681H  replacement is in the binding domain [8]. N501Y replacement probable affects binding also, as it is near the binding site. It is also found in the South African rapid-spreader; but by a different change in the RNA, so it arose there indepencently. 

            There are 5 other substitutions in Spike, and two small deletions. The deletion 69-70 causes one of the PCR diagnostic probes for covid-19 (ThermoFisher TaqPath probe) to come up negative [2], though other primer pairs can still identify the presence of  SARS-CoV-2 RNA. The group of Volz et al.[2] tentatively use the failure to detect covid with that probe as a surrogate identifier for the VOC 202012/01, saving them the need to sequence the genome. Using that assumption they conclude that VOC is over-represented in the younger (0-20 yrs) cohorts of COVID-positive patients. 

            Till now (2021-01-02), no one has found any evidence of a more aggressive disease with the variant viruses. 

            It is estimated that VOC 202012/01 is 56% more transmissible than preexisting SARS-CoV-2 [9]. Under a regime where the R₀ was 1.0, it might rise to 1.56.  There is as yet little evidence as to why the variant seems to have a higher reproduction rate; possibilities include (a) tighter binding of mutant spike to human ACE-2 receptor requiring lower titre for infection, (b) shorter lag between infection and shedding, (c) longer survival of infective RNA in the air or on surfaces, (d) evasion of RNA hydrolysis inside cells.  And doubtless others.

References

[1]  https://www.ecdc.europa.eu/sites/default/files/documents/SARS-CoV-2-variant-multiple-spike-protein-mutations-United-Kingdom.pdf

[2]  https://www.imperial.ac.uk/media/imperial-college/medicine/mrc-gida/2020-12-31-COVID19-Report-42-Preprint-VOC.pdf

[3] Science  18 Dec 2020:Vol. 370, Issue 6523, pp. 1464-1468 DOI: 10.1126/science.abe8499

[4] Cell. 2020 Oct 29;183(3):739-751.e8. doi: 10.1016/j.cell.2020.09.032. Epub 2020 Sep 15

[5] Nature 2020 Mar;579(7798):265-269. doi: 10.1038/s41586-020-2008-3. Epub 2020 Feb 3. https://pubmed.ncbi.nlm.nih.gov/32015508/

[6] https://www.ecdc.europa.eu/en/publications-data/covid-19-risk-assessment-spread-new-sars-cov-2-variants-eueea

[7] https://www.bmj.com/content/371/bmj.m4857

[8] https://www.who.int/csr/don/21-december-2020-sars-cov2-variant-united-kingdom/en/

[9] https://doi.org/10.1101/2020.12.24.20248822

 

 

 

 

RNA Vaccines

Currently (12 Nov. 2020), the World Health Organisation (WHO) is aware of 

48 different teams around the world who are working on the production of vaccines against the SARS-Cov2 virus that have already got to the stage of clinical evaluation. The Pfizer/BioNTech/FosunPharma team, known to all since 9th November when it announced a degree of success, is one of these 48. (There are in addition 160 other vaccine-production teams that are in pre-clinical stages of development.)

A number of different vaccine technologies are being tried in the 48 different vaccine-teams that are already at the stage of clinical trials, listed by WHO [1].  I summarise these below. 

Type 1. Inactivated virus (in this case inactivated SARS-Cov2). This is the approach used in the Salk polio vaccine, which used formaldehyde to ‘kill’ the virus. Formaldehyde can modify the shape of proteins, and the antibodies produced may only react with formaldehyde-treated virus. The virus must be really, really, dead, and safety is a perpetual concern.

Type 2. Replicating viral vectors. Thus, the SARS-Cov2 spike protein gene can be inserted into the genome of a mild virus (e.g. the measles virus or adenovirus). These viruses have their own way of getting into cells and replicating, but introduces a SARS-Cov2 antigen, against which the host can raise antibodies. (See [2]) 

Type 3. Non-replicating viral vectors. Adenoviruses often used.These can get into cells but will not spread in the host. Higher doses are therefore needed. (This is the strategy used by the Oxford/Astra Zeneca team.) 

Type 4. DNA vaccines. The mRNA for a viral gene is copied (using reverse transcriptase) into a double-stranded DNA plasmid that grows happily in bacteria. Large quantities of the plasmid are grown up, purified on columns and used as vaccine. Once inside a cell they should direct the synthesis of e.g. Spike protein (amongst several others.)

Type 5. Protein subunit vaccines. These are a more recent development, and becoming popular, as no virus is involved in the manufacture. The gene for a viral protein can be used to produce large quantities of the protein in vitro. However, the isolated and purified protein may not have the right shape to trigger formation of antibodies effective against native virus. 

Type 6. Virus-like Particles (VLPs) can be prepared by growing cultured cells that produce only sufficient of the viral proteins to form a particle, but are not able to reproduce whole virus. If RNA is needed to form a particle, small bits of irrelevant RNA can be added. These particles, purified from cell cultures, can be used as vaccines, and are often more potent antigens than the isolated soluble protein or protein subunits of type 5. Again, no virus is involved in the process of manufacture.

Type 7. RNA vaccines. In this strategy single-stranded mRNA that codes only one viral protein (e.g. the Spike protein) is encapsulated in a Lipid Nano Particle (LNP) some 70-100 nm in diameter [3] (1million nm = 1 mm). Human cells have an inherent tendency to engulf particles of a particular size and attempt to digest them (a hangover, no doubt, from our amoeboid ancestry). The released mRNA directs the synthesis of spike protein (or its Receptor Binding Domain) in the cell. This technology has been developed over the last 20 years for experimentally silencing genes; and since 2012 for producing vaccines against single-strand RNA viruses such as influenza. It was first used in humans in 2017 [4]. The advantage is that an equipped factory can turn to producing a novel vaccine within a week. All it needs is to know the sequence of the mRNA. (This is the strategy use by the Pfizer/ BioNTech/ FosunPharma team, and a team at Imperial College, London.) RNA is far more susceptible to hydrolysis than either protein or DNA (because of the -OH, group missing in 2' desoxyribose). Vaccines are conventionally kept at 5-8ºC, but RNA vaccines must be kept at –78ºC or lower. That is not a problem. A 6 litre dry-ice or liquid nitrogen Dewar, twice the size of a pressure cooker, will hold its temperature for 200 days.

We see that the different vaccine strategies have their own advantages and disadvantages. The RNA technology has the advantage of speed; and relative safety. 

References

[1]    https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines

[2]    https://doi.org/10.1016/j.virol.2014.01.002 

[3]    https://www.liebertpub.com/doi/full/10.1089/nat.2018.0721

[4]    https://pubmed.ncbi.nlm.nih.gov/28457665/

SARS-CoV2 (Continued)

SARS-CoV2 (Continued)  

     One of the puzzling features of this virus is that some infected people, while carrying a considerable viral load, and shedding infectious virions, nevertheless develop such trivial symptoms that they never notice they are infected. While other, of course, develop virally driven hyper-inflammation,  respiratory failure, and sometimes also kidney and heart failure.

     This variability in response is especially striking when it affects a whole nation-state. Thus the official WHO figures record that Vietnam, with a population of 97 million, has confirmed only 416 cases of COVID-19, and that none have died. (c.f. UK, population 67 million, 297,914 cases, 45,677 deaths on 24th July. )

Possible explanations for variable responses.

     The hypothesis of genetic variability in the human host, which supposes that the Vietnamese lack e.g. the ACE2 receptor site (See my SARS-CoV2 post), is more-or-less ruled out by anecdotal observations such as that of an asymptomatic carrier infecting 5 family members [1]. As also is the hypothesis of genetic variability among the circulating SARS-CoV2 strains, for the carrier would obviously infect the household with the strain she was carrying. 

     Could there be competition between two co-infecting viral strains, where one causes trivial, often negligible, symptoms but occupies all the binding sites? 

     Or could there be, in some people, residual anti-bodies at a sufficient titre from a previous infection by the same (or sufficiently similar) coronavirus?  This last seems the best hypothesis, and in the last 10 days has received some support. 

    The group of Antonio Bertoletti at the Duke-NUS Medical School in Singapore has just published in Nature [2] an online report showing that previous infection with a virus of the beta-coronavirus family can leave long-lasting and multispecific T cell immunity to the nucleocapsid structural protein (N protein, or NP, See my Coronavirus post) that can cross-react with the N protein of SARS-CoV2). This previous infection could be a harmless "common cold" member of the corona virus family, but in Singapore it was possible also to study survivors of the 2003 SARS pandemic. 

     Back in 2013 a group in Taiwan explored the antigenicity of the N protein of the mild common cold virus HCoV-OC43, and had found that the middle section was highly antigenic [3]. Well over 90% of healthy young adults contained antibodies in their serum against the N protein of this common virus. These antibodies were even found in cord-blood samples showing that newborn babies acquire some immunity against coronaviruses from their mothers. 

     We have already learnt that it is foolish to infect yourself deliberately with SARS-CoV2; you could become very ill or die. But there may be a beta-coronavirus, prevalent in Vietnam, that does protect you against COVID-19. And it may be that here in Britain a sufficiently recent 'common cold' may leave you with enough circulating antibodies to prevent or greatly limit the effect of SARS-CoV2 infection. 

References

[1]  Susan Lee,  Paula Meyler,  et al. (2020) Can J Anaesth. : 1–7. "Asymptomatic carriage and transmission of SARS-CoV-2: What do we know?". 

[2]  Le Bert N, Tan AT, Kunasegaran K, et al. (2020)  "SARS-CoV-2-specific T cell immunity in cases of COVID-19 and SARS, and uninfected controls." [published online ahead of print]. Nature. 2020;10.1038/s41586-020-2550-z. doi:10.1038/s41586-020-2550-z.  

See also  Ruairi J. Mackenzie, Science Writer for Technology Networks (2020) “'Common Cold' Coronaviruses Could Help Produce Anti-SARS-CoV-2 Immune Cells." 

[3]  Fang-Ying Liang, Leng-Chieh Lin, (2013  J Virol Methods;187(2):413-20. "Immunoreactivity characterisation of the three structural regions of the human coronavirus OC43 nucleocapsid protein by Western blot: implications for the diagnosis of coronavirus infection. "

SARS-CoV2

SARS-CoV2

    After looking briefly into coronaviruses in general (see previous post), I have turned to SARS-CoV2, the virus responsible for the current pandemic of the respiratory infection called COVID-19.

Interesting aspect at this stage in the pandemic include:

• Important biochemical and biological differences between SARS-CoV2 and SARS-CoV viruses.
• Hight infectivity; 100 or 1000 times higher than SARS-CoV.
• Why are 30-40% of infected  'carriers' symptom-free?

Difference between SARS-CoV2 and SARS-CoV(1) that might cause higher infectivity.

    SARS-CoV2 is said to be 1000 time more infectious than SARS-CoV1; a pretty loose statement, but there is some biochemistry to investigate. Is this high infectivity due to:  

a.    different, more accessible or numerous, targets on host;

b.    tighter binding to target; 

c.     epidemiological factors like shedding before or without symptoms, or more coughing and sneezing; 

d.    better evasion of host responses. 

[a] Target on host

    The host receptor for both SARS-CoV and SARS-CoV2  seems to be the dimeric membrane-bound protease called ACE2 (for Angiotensin Converting Enzyme 2). There is a small mystery here, as the first investigators found very little ACE2 protein, or mRNA, in human lung tissue, though lots in arteries, gut, kidney, testes and elsewhere [1]. Yet SARS-CoV2  seems to attack the lower respiratory tract (as well as gut, blood-cells, kidney, etc). This was so important that the question was re-examined and some ACE2 was found in lung tissue, particularly around arterioles [2].  (It is notable, however, that SARS-CoV2  can cause diarrhoea and kidney damage [3].) Other coronaviruses act primarily as gut pathogens (See previous blog). I worried that the polyclonal antibody used by Hamming to test the presence of ACE2 (which was reared against a stretch of 19 amino acids distinctive to ACE2) might cross-react and mislead. However, it seems to be universally accepted that the receptor for SARS-CoV2  is ACE2.

[b] Tighter binding of Spike to Target.

    The spike protein is not highly conserved; the opposite rather, and it seems likely that mutations, deletions and insertions in spike can affect host range; possibly infectivity as well. There are distinctive features in the spike protein of SARS-CoV2 not found in SARS-CoV spike.  Thus, there is an insert of 4 amino acids (PRRA) into the sequence, which generates a cleavage site absent from the spike protein of SARS-CoV and several other coronaviruses (but present in MERS!). 
                                                                                ↓            
            SARS-CoV2: CASYQTQTNSPRRARSVASQSI
            SARS-CoV  : CASYQTQTNS­­– – – –RSVASQSI
Cleavage is effected by a host protease present cytoplasmically throughout the body. It is a 'subtilisin-type' calcium-dependent protease (called furin), which cleaves after the marked serine residue, but the cleavage site is flagged by the paired basic amino acids (–R+R+–). The furin enzyme is obviously present to service host proteins. But SARS-CoV2 is not unique among viruses in using it for pathologicial purposes, for furin also operates in the activation of: HIV, influenza, dengue fever, Marburg virus, papillomavirus, and even anthrax toxin. It is suggested that processing of progeny virions before release may facilitate the spread of virus (c.f. SARS-CoV)[4]. 
    The spike protein of SARS-CoV2 operates as two peptides (S1 and S2) formed into a trimeric "clove-like" structure.  Tai et al. (2020) were able to compare the binding (to human ACE2) of SARS-CoV2 spike with that of SARS-CoV spike, and found it bound 9 time more tightly. (Interestingly, it bound even more tightly to bat ACE2). [5] 

[c] Epidemiological factors.

     It is important to distinguish pre-symptomatic from truly a-symptomatic carriers; both categories of infected subjects experience no symptoms, but in the former case they eventually develop symptoms, while in the latter they never do. When I use the term 'asymtomatic' in this post it will always mean that the subject did not develop sysmptoms, in at least 4 weeks. Both categories can spread the disease.
    Compared with SARS-CoV of 2003, SARS-CoV2 causes more cases with mild (or very mild) symptoms, and larger numbers stayed at home in the community. There was also twice as long incubation period before the appearance of symptoms (4–12 days). Similarly relevant for the spread of the disease, the new strain can shed infective particles as soon as symptoms appear; or even before (see above). And they can continue shedding for 3 weeks [6].   Susan Lee et al. [7] mentions a family in Anyang (China) where an asymptomatic carrier tested positive for the virus and infected 5 family members. 
    Other factors of obvious relevance to infectivity are propensity to cough or sneeze. 

[d] Evasion of host defences.

    Two recent well referenced summaries are by Indwiani Astuti and Ysrafil [8] and Swatantra Kumar et al. [9]

 References

[1] Donoghue, M., Hsieh, F. et al. (2000) Circulation Research. 87:e1–e9; "A Novel Angiotensin-Converting Enzyme–Related Carboxypeptidase (ACE2) Converts Angiotensin I to Angiotensin 1-9".
[2] Hamming I, Timens W, Bulthuis ML, Lely AT, Navis G, van Goor H.  (2004)"Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis." J Pathol.; 203:631-637. doi:10.1002/path.1570
[3] Martinez-Rojas, M.A. et al. (2020) "Is the kidney a target of SARS-CoV-2?"; Am J Physiol Renal Physiol.; 318:F1454-F1462.
[4] Coutard, B., Valle, C. et al  (2020)
Antiviral Res.; 176: 104742. "The spike glycoprotein of the new coronavirus 2019-nCoV contains a furin-like cleavage site absent in CoV of the same clade."
[5] Tai, W-B. He L.,  Zhang, X-J. et.al. (2020) Cellular & Molecular Immunology volume 17, 613–620. "Characterization of the receptor-binding domain (RBD) of 2019 novel coronavirus: implication for development of RBD protein as a viral attachment inhibitor and vaccine"
[6] Petersen, E., Koopmans, M. et al (2020).Lancet, Infectious Diseases, https://doi.org/10.1016/ S1473-3099(20)30484-9 "Comparing SARS-CoV-2 with SARS-CoV and influenza pandemics.".
[7] Lee S, Meyler P, Mozel M, Tauh T, Merchant R. "Asymptomatic carriage and transmission of SARS-CoV-2: What do we know?"  Can J Anaesth. 2020;1-7. doi:10.1007/s12630-020-01729-x
[8]  Indwiani Astuti & Ysrafil, (2020) Diabetes Metab Syndr. 2020 July-August; 14(4): 407–412.
Published online 2020 Apr 18. doi: 10.1016/j.dsx.2020.04.020 "Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2): An overview of viral structure and host response"
[9]  Kumar S., Nyodu R., Maurya V.K., Saxena S.K. (2020) Host Immune Response and Immunobiology of Human SARS-CoV-2 Infection. In: Saxena S. (eds) Coronavirus Disease 2019 (COVID-19). Medical Virology: From Pathogenesis to Disease Control. Springer, Singapore. Published online 2020 Apr 30. doi: 10.1007/978-981-15-4814-7_5